UK: gambling – the myth of growth


UK: gambling – the myth of growth

“Without continual growth and progress, such words as improvement, achievement and success have no meaning” Benjamin Franklin
As with most businesses and sectors, gambling companies like to talk about growth. For those that are listed, there is an obvious reason why; but even private companies find a growth narrative far more comfortable than a nuanced view of the top-line: it makes spending decisions simpler, it is easier to attract talent and it provides a confident framework for future prospects. For most businesses, most of the time, growth is a good thing all round. But gambling (in the UK and elsewhere) has fallen into something of a trap: it has talked up growth, become much more visible, and increasingly attracted the ire of key opinion formers (press, TV documentaries etc) as well as – more problematically – lawmakers. There is an ironic twist to Britain’s gambling boom, however: it hasn’t happened. Gambling has been in relative decline on a generational view, only recently picking up a modicum of real growth – which in turn has now evaporated. Put simply, a materially greater proportion of GB Household Disposable Income (an ONS figure measuring what consumers have to spend) was spent on gambling 1999-2006 (1.10%) than in any year thereafter (0.96% 2008 – 2012; 1.04% 2013 – 2018: which points to cyclicality as well as regulatory impact). Gambling isn’t booming, but the industry’s self-image of growth might just have blown it anyway…

Of course the UK gambling market has grown in the period between 1999 to 2018. In fact it has grown by £7.0bn, or 85% – which sounds like a lot. However, there are two important caveats to this data. The first is basic but often misunderstood by wider stakeholders. The GB Gambling Commission only started fully regulating landbased gambling (and lotteries) in 2007 – when officially collated revenue records began (though a combination of listed company data, Companies House filings and HMRC/HMT data allows us to build a pretty accurate picture of what went before). But the Gambling Commission only  started domestically regulating online gambling for a part period in 2014-15, which is when online data is added to the official industry statistics. So if we compare total official revenue in 2007-8 with total official revenue for 2017-18 we are comparing landbased only to landbased + online. In 2008, online was already a c. £1.35bn revenue market (bigger than bingo and casino). So in the period 2007-8 to 2017-8, the UK gambling industry did not grow by 76% (5.8% CAGR), but by 52% (4.3% CAGR). Those that prefer a good story to solid figures might think we are splitting hairs here – both look like big numbers. But one isn’t.

Growth of 52% over ten years sounds like a lot. But the economy has also grown a lot over ten years. Household Disposable Income growth over the same period was 36% or 3.1% CAGR. The gambling industry over the last decade has therefore outstripped consumer spending power by a material but not all that significant 1.2ppts per annum. In other words, the proportion of disposable income that British consumers are spending on gambling has moved from 0.95% in 2008 to 1.06% in 2018 – hardly an explosion, even factoring in a (lottery-led) reduction in the overall player-base.

But even this is not the complete story. 2008 was the last ‘good year’ for the British consumer before the financial crisis became widespread recession: in 2009-11, HDI growth fell from a previous average of 4.5% p.a. to just 1.9%: materially below the prevailing headline rate of inflation (3.0%) and so a real-terms decline in living standards. 2008 was not that great a year for gambling, however (other, notably, than bookmakers – likely benefiting from substitution). Landbased gaming sectors were still reeling from the Smoking Ban (H2 2007) and the loss of S16/21 machines (similar to FOBTs; Q4 2007): the aggregated revenue change for casinos, bingo, arcades and machines in pubs was -13% or nearly £500m lost (i.e. a shock broadly similar in magnitude to the B2 ban given inflation, but more widely felt). Growth was even spluttering somewhat online (up ‘only’ 8.5% in 2008), in the gap between heavy user adoption of a relatively poor desktop product (with limited to no mass market appeal) and the broadband / in-play / mobile / advertising growth drivers that powered the sector from 2010 onward. Therefore – and optically very importantly – the official data (April 2008-9) starts at a trough in overall gambling spend combined with a cyclical peak in consumer spend.

If we go back another decade, we find overall gambling spend averaged a fairly resilient 1.1% of HDI, before the public health and regulatory changes of 2007 uncoupled this and brought it down to 0.91% (2009 was the trough: negative regulatory change + recession = bad news for gambling). Therefore, what online has done between 2010 and 2018 is not grow gambling to new highs, but allowed it to normalise some way back to pre-recessionary and regulatory impact (which, significantly, were not designed to cut gambling demand levels at all). Indeed, to get back pre-recessionary equivalent levels of consumer spend / sector revenue, gambling would need to add another £1.4bn – an additional nearly 10% of the current industry total, and then grow with at least inflation.

We have included lottery in the total since regular draw-based and scratchcard consumption is not all that dissimilar from the occasional flutter in value or frequency terms, while very regular multiple draw-based participation and scratchcard use is at least fungible with a Saturday acer and a few in-play bets on big games/races or a couple of trips to the casino/bingo hall: the extent of most people’s gambling involvement, if not the source of most revenue. Nevertheless, there is a case to be made for lottery being ‘different’ on a number of levels (not run commercially, core draw-based product specifically not built around rapid play, much higher tax and good cause elements). However, if we strip out lottery and consider only commercial gambling, the picture doesn’t change all that much: from 1999 to 2006 the average proportion of HDI spent on commercial gambling was 0.79%, peaking at 0.82% in 2005 and 2006; from 2009 to 2018 it was 0.73%, reaching 0.82% again in 2017, but now in relative decline again (from a recessionary low of 0.66% 2009-2010) – so the same pattern of land-based regulatory and recessionary relative decline, followed by a decade to recover – but to only pre recessionary levels (and commercial gambling spend in 2019 will be some way off 2017 relative highs, given likely absolute revenue decline, causing a fall back down to c. 0.75% – see below).

For the UK, 2017 therefore represented relative peak commercial gambling spend since the 2009 recession, itself likely helped materially by very high sportsbetting margins in Q4. However, even this high only represented a return to pre-recession and (accidental demand impact) regulatory change and is now falling again. Indeed, 0.75% would put the sector right back at its long-term average for 20 years: no real terms growth in relative consumer spending achieved at all.

There has not therefore been an explosion in gambling growth; instead, gambling is £1.4bn light of where it should be to get back to pre-recessionary levels of revenue when factoring in lottery decline, and was only just back at pre recessionary levels excluding lottery (though now facing a c. £1.0bn headwind from cumulative regulatory changes: VIP-related and B2 ban). This total ‘real revenue gap’ figure quite a big number: bigger, for example, than the landbased bingo or casino sectors, more than the total gambled on football or horseracing. But there are logical reasons for this gap occurring, in our view:

  • Lottery participation: lottery participation has been falling relatively consistently for the period in question (especially 2000-8 and 2013 – 18) and this has inevitably had an impact on sales: overall lottery revenue (including Large Society Lotteries such as Peoples Postcode and Health)  has grown by only 1.5% pa in the period, with the National Lottery growing by only 0.7% pa : this is a significant drag on overall performance given that lottery started out with a 33% share of gambling spend and is now down to 23% – absolute growth can still mean significant real terms decline. However, as we explained above, this is an exacerbating factor, but not the cause
  • Retail development: such as there was, largely occurred in the 1990s (bingo) and early 2000s (entertainment-led, large format casinos; LBOs with roulette); the retail firmament was largely set by the time of the recession (with the arguably rather narrow exceptions of SSBTs, electronic bingo terminals and digital slot machines) – fifteen years of limited to no R&D, an almost accidental legislative kicking midway through, channel shift and a wider high-street footfall malaise meant that retail gambling in the UK has managed annual growth of only 0.9% over twenty years: real terms decline of 3% pa and no underlying recovery from the 2009 recession (it is worth pointing out that the casino sector has broadly kept up with consumer spend growth at 3.7% CAGR over the period, but given the regulatory and operational state of the British casino industry in 1999, this is hardly surprising – its ability to grow strongly going forward is much more open to question, in our view)
  • Recession: before 2009 gambling executives loved to say that gambling was defensive, almost recession-proof. The experiences of 2009 modified that view to ‘resilient’. Resilience is relative, but gambling needs to be a consumer discretionary if it is to have broad social acceptability (that the social responsibility corollary of the notion that people would gamble more during hard times did not occur to champions of the myth says a lot about the state of the industry a decade ago). Fortunately for the long-term sustainability of the sector, if potentially not for the next couple of years, gambling is cyclical insofar as it moves pretty closely with HDI – it is not terrifyingly cyclical (like high-end restaurants or owning racehorses), but it is not particularly defensive either (correlation between HDI and gambling revenue is a convincingly high 0.96). The big problem with the 2009 recession, however, was not that that it reset spending, but that the sector was only firing on one cylinder – online – coming out of it
  • Regulatory change: the years 1999-2004 were periods of considerable hoped-for rebirth for the British gambling sector, with all eyes on the potential for a new liberal act and what that would do to landbased gambling; the slow motion car crash that happened next (with implementation followed by the smoking ban) was still having negative (for operators) regulatory repercussions up to April this year (and still has for 68 Act casino operators, those stuck with archaic machine categories or the poorly codified imbalance between landbased and online product/consumer protection capabilities); added to this backdrop has been a new regulatory focus on VIPs (casinos, especially London, and online) which has likely cost a c. -£400m swing in sector revenue (some already in 2018 figures). Whether this revenue can or will come back is an open question, but it is certainly a big factor in the current growth stall. Added to this is the further c. £700m net-lost (though potentially part redistributed) due to the B2 ban, which is not captured in the historical data. Due to these regulatory changes, overall gambling is likely to have a tough 2019 and almost certainly absolute overall revenue contraction (bigger in absolute terms than 2007-9) regardless of (waning) consumer resilience
  • Online engagement dynamics: on a 20 year view online has clearly been the star performer, with an overall CAGR of 16.2% from relatively infinitesimal beginnings before the Millennium to representing over half commercial sector revenue (not magicking into existence in 2014); however, online growth has now slowed to single digits despite potential room for further growth (that theoretical £1.4bn of additional potential spend is logically mostly online), part of this is regulatory (VIPs, see below), part is also likely to be a pause from the super-normal growth driven by smartphone adoption (which is now ubiquitous). But a big part in our view is that the vast majority of the online gambling sector is still pointing to the heavy user (with busy UX and brand-free CPAs of £100+), which cannot economically monetise the mass market (who are often put off by the product and anyway don’t want to spend enough quickly enough to make a £100 CPA justified) – with the arguable exceptions of the retail brands (some structurally low CPA + cross-sell), Sky, Tombola and InTouch (with routes in pub slots), most operators (or the supply chain) does not have an answer to this problem, which essentially retards growth because it is uneconomic to achieve (especially if the regulator is also watching). Reductions in advertising are probably double edged here: on the one hand reducing the ability for operators to reach mass markets, but on the other reducing the ability of bad operators to waste money. Crucially, the retail sector doesn’t have this problem: occasional footfall is typically high margin spend on a fixed cost base, but given the issues highlighted above, this footfall is dwindling
On a longer term view, gambling in Britain has not been growing in real terms: the industry had barely been able to claw back some of its pre-recessionary position before a new wave of regulatory change has driven it back further. There are some structural reasons for this, some regulatory but also plenty that are strategic/operational. The big question for the UK gambling industry, therefore, is not whether it can maintain its pace of growth under increased regulatory pressure, but whether it can responsibly engage with a wider audience again and sustainably build in reality the overall growth story it thought it already had…